Steel/Copperweld antenna wire...losses?
Reg Edwards
manufacturers ever specify RF losses or other electrical characteristics.
They are not equipped to measure or guarantee such parameters. They are in
the wire drawing metal industry. But they will provide dimensions,
mechanical and metallurgical properties of their products. There may exist
one manufacturer who does provide electrical data.
But armed with the diameters and thickness of the copper (coating,
electroplating, drawing, extruding process) with the aid of a pocket
calculator you can very easily work things out for yourself. Put the
following data in your technical notebook -
The skin depth in copper is 2.6 thousandths of an inch divided by the
squareroot of the frequency in MHz.
In metric terms, the skin depth in copper is 0.0661 millimetres divided by
the squareroot of the frequency in MHz.
When the copper layer (over the steel) is less than twice skin depth, the
RF resistance of the composite conductor will begin to be measurably higher
than that of a solid copper wire of the same overall diameter. Very little
current will flow in the steel core, not only due to its higher resistivity
but also due to its much higher permeability and inductive reactance. So we
are left with a very thin-wall copper tube.
For example, at 1.8 MHz the skin depth in copper at 20 degrees C is 0.00194
inches. So ideally, the copper thickness should be at least 0.00388 inches.
Thats pretty thick !
But much depends for what application the copper-over-steel wire is
intended.
Another example - The effective RF resistance of a solid copper, 1.6 mm
diameter wire, half-wave dipole on the 160 meter band is 2.0 ohms. The
radiation resistance is 73 ohms. So the radiating efficiency is 73/75 times
100, equal to 97.3 percent.
If the solid copper wire antenna was replaced by a "copperweld" conductor
having an RF resistance of THREE times the previous value, the radiating
efficiency would fall only to 92.4 percent. The difference in received
signal power of 5 percent between the first and second antennas would be
competely undetectable - about 1/28th of an S-point.
The prospects of obtaining any useful info based on practical experience is
remote. So find the thickness of copper over a steel core from the
manufacturer and work it out for yourself.
--
Reg G4FGQ
http://www.btinternet.com/~g4fgq.regp
Tom Bruhns
Reg Edwards (g4fgq...@btinternet.com) wrote:
: But armed with the diameters and thickness of the copper (coating,
: electroplating, drawing, extruding process) with the aid of a pocket
: calculator you can very easily work things out for yourself. Put the
: following data in your technical notebook -
: The skin depth in copper is 2.6 thousandths of an inch divided by the
: squareroot of the frequency in MHz.
: In metric terms, the skin depth in copper is 0.0661 millimetres divided by
: the squareroot of the frequency in MHz.
: When the copper layer (over the steel) is less than twice skin depth, the
: RF resistance of the composite conductor will begin to be measurably higher
: than that of a solid copper wire of the same overall diameter. Very little
: current will flow in the steel core, not only due to its higher resistivity
: but also due to its much higher permeability and inductive reactance. So we
: are left with a very thin-wall copper tube.
: For example, at 1.8 MHz the skin depth in copper at 20 degrees C is 0.00194
: inches. So ideally, the copper thickness should be at least 0.00388 inches.
: Thats pretty thick !
Of course, Reg's example is pretty much a "worst case" in that higher
frequencies have shallower skin depths, and therefore are less affected
by the steel core. My books suggest that copper clad steel wire in "30%
conductor" and "40% conductor" varieties are the most common
commercially. (BTW, "Copperweld" is a trademark, I believe.) OK, some
calcs:
Wire AWG Wire Copper layer thickness
Gauge Diameter 30% cond. 40% cond.
number inches inches inches
10 0.1019 0.0083 0.0115
12 0.0808 0.0066 0.0091
14 0.0641 0.0052 0.0072
16 0.0508 0.0041 0.0057
18 0.0403 0.0033 0.0045
Assmuing no math or transcription errors in the above numbers, it's only
the 30% conductor 18 gauge wire that fails to meet Reg's recommended
minimum copper thickness. Just how well the manufacturers maintain
their processes so that the copper cladding really is a uniform
thickness along the length and around the circumference of the wire, I
couldn't say. But in any event, 12-gauge copper clad steel should be
fine for antennas in the ham bands.
I once used a 20m dipole made from thin stainless steel aircraft control
cable, and I can tell you that it did seem to be less efficient than I'd
have liked. But stainless has pretty low conductivity. At least it stayed
up in all sorts of nasty winds. Anyone know if copper clad stainless
is available?
--
Cheers,
Tom
tom_b...@hp.com
Roy Lewallen
this is an appropriate place to mention something I ran into with
copper plated wire. I once got some small-diameter (about 0.1") 75 ohm
coax, and discovered that its loss was 4 dB/100 ft at 7 MHz, much
higher than I expected. The reason was that the center conductor was
made from several strands of very small diameter copper clad steel
wire. Although the plating was apparently quite thick as a percentage
of the overall diameter, it was still thin enough that significant
current flowed in the steel, resulting in high loss. At higher
frequencies, the loss eventually got to be what you'd expect from
copper. Some RG-174 I've used also began showing this problem in the
lower part of the HF range, for the same reason. Apparently, copper
clad steel is commonly used for small diameter coax for strength. With
very small diameter strands, the copper can be too thin at lower
frequencies.
Roy Lewallen, W7EL
Jerry Flanders
conventional stranded galvanized steel tower guy cable (1/4 or so inch
dia) instead of copper for a dipole? I will be putting up a 70 ft Rohn
25 tower next spring and would like to get double duty from the guy
system as a pair of inverted vees.
Jerry W4UK
On Tue, 25 Aug 1998 01:07:07 GMT, Roy Lewallen <w7...@teleport.com>
wrote:
Robert Lay (W9DMK)
Hi Tom,
The issue of whether or not the CopperWeld copper plating is thick
enough or not is interesting, but as I read the posts from Reg and
yourself, I wondered about something else.
Does the concentration of current on the outer surface of a solid
copper conductor still distribute itself in the same way when the
inner core of copper is removed entirely? In other words, does the
same ac resistance obtain from a thin walled copper tubing of X
diameter as would obtain from a solid copper rod of X dia?
Obviously, if there is a difference, then the inner core being steel
with its higher resistivity may give you some unexpected results.
Tom Bruhns
: Hi Tom,
So long as the copper tube wall is several "skin depths" thick,
the tube and the solid conductor should conduct the same. In
fact, you can wind coils from copper tube and run water thru the
tubing to keep a coil cool... You can even use the tube to
supply water to your water cooled power amplifier. Welllll...that's
drifting a bit far off-topic, I guess!
As an example, if the copper tube wall is 5 skin depths thick,
the current at the inside wall should be about 0.7% of the current
at the outer surface. So adding a solid copper core would
decrease the resistance, in that case, by less than 1%. Detectable,
but probably not an issue in most ham work. (I'd be more worried
about the surface irregularities on the OUTSIDE of the tube/wire.)
I suppose the current distribution in the outer section is affected
somewhat by the presence or absence of a core, if that core carries
current, but I'd expect that effect to be quite small. Maybe an e&m
expert will jump in and fill in the details more accurately!
--
Cheers,
Tom
tom_b...@hp.com
Reg Edwards
magnetic field at any conductor radius smaller than its own. It has no
effect on the core material.
The magnetic field set up by current flowing in a conductor is entirely
external. To apply inductive loading to a wire the magnetic material must
be outside. Eg, a ferrite sleeve must be placed OVER the wire.
So if the copper layer is thicker than about two skin-depths, no current
flows in the steel core and there is no magnetic field in the core. All
losses are confined to the copper.